Probably the easiest way to see this is to say that if you want to make domestic policy more left-wing than it currently is, what you really need to do is to change congress and especially the Senate—replace some incumbents with people who are more liberal. On national security policy, this really isn’t the case. If you replaced Judd Gregg with Paul Hodes, that wouldn’t change our policy in Afghanistan it wouldn’t alter the kill order on Anwar al-Awlaki, it wouldn’t change the drone strikes in Pakistan, it wouldn’t alter the pace of withdrawal from Iraq, etc. By contrast, if you replaced Barack Obama with someone who had deeply held left-wing views on these subjects things would change. That hypothetical president wouldn’t operate without constraint, but would have substantial latitude to change things especially on the topics he cared most about and was willing to suffer public criticism over.I have a two part response on this issue. What I'll talk about in this post is, more or less, about "normal" politics when it comes to national security and foreign affairs. That's pretty much what Yglesias is talking about here. There's a second question, of what might be thought of as extraconstitutional politics in national security...I'll have to save that for a second post.
Okay, so...regular politics. If you replace Gregg with Hodes, it's true that it doesn't change the pace of withdrawal from Iraq. But I do think that replacing a Republican majority in 2005-2006 with a Democratic one in 2007-2008 made quite a bit of difference. The policy of happy talk and drift that Bush followed after the 2004 election was no longer sustainable after the Democrats moved into majorities in Congress. Granted, the immediate perverse consequence was the surge, which certainly wasn't what antiwar voters wanted. But over the course of the two years, what happened was cashiering Rumsfeld, replacing drift and happy talk with a much more actively managed war, and then
Meanwhile, given the current distribution of the parties in Congress, it's easy to see how Hodes/Gregg would have made a significant difference on many issues...expand that to a couple of other close races, over the last three cycles, and certainly adding three mainstream Dems and subtracting three mainstream Republicans would change outcomes in the Senate. That said: not as much as swapping Obama for John McCain in the White House! Indeed, while I do think that Obama's "choice" of health care was severely constrained by coalition politics within the Democratic party, it's not hard to imagine a slightly different world in which the faction that placed a higher priority on climate/energy won the nomination over the faction that cared more about health care.
That's all pretty scattered; let's see if I can narrow it down a bit.
First. Changing the president is always going to matter more than changing any single Member of Congress. No one doubts that the president is more important than individual Members of Congress (I'm sure Yglesias agrees that this is true in domestic policy, too).
Second. Lots of things that seem to be choices by the president really are not. Take, for example, Afghanistan. Again, it sure seems to me that doubling down there was the consensus liberal foreign policy establishment position in the 2008 primaries; did anyone (other than out-of-the-mainstream figure Dennis Kucinich) have a different position? Certainly, there are plenty on the left who wanted out of the area entirely, but that's not what Obama (or Hillary Clinton) ran on. Second, it seems fairly clear that the generals running the war there wanted to increase the US presence. Third, outside of Ron Paul and a handful of people who support him, most of the partisan opposition in Congress were for something that looked surge-like (in 2009, if not, you know, when Republicans were actually in a position to do something about it). Doing something that could be sold as "doing what the generals proposed" was clearly the path of least resistance. Was it also what Obama really wanted? I don't know; what I do know is that it's a mistake to assume that because he eventually supported it, it must have been his first choice.
Third. As I said in one of the previous posts, it is certainly true that the president's formal powers are in some ways stronger in national security policy, and that does help him in those areas. As commander in chief, he does have an asset in battles against the military that he does not have in battles with civilian agencies (although, as Yglesias points out, generals tend to have more prestige than do other bureaucrats). I think it's fair to say that in general, Members of Congress probably care more about domestic policy than they do about military and foreign affairs policy -- although there are tons of exceptions, whether it is Jewish and Cuban constituencies who care about policy in a particular part of the world, or Members who are responsive to the bases or military contractors found in their districts. In cases in which few if any constituency groups care, it's true that the president will have fewer constraints.
Fourth. However, for the same reasons that Members of Congress may not have strong positions, it may be the case that the president doesn't have strong preferences. This is pure speculation, but that's my guess about explaining Barack Obama's failure to close the prisons at Guantanamo Bay. I think the evidence is that Barack Obama's first choice was in fact, as he stated, to close Gitmo. But it probably wasn't a strong preference...not in terms of it seeming a close call to him, but in terms of where it landed on his list of priorities. So when people pushed back, he didn't have the same incentives to fight on as he had, for example, in health care. That doesn't mean it shouldn't count as a defeat for him (as opposed to foolishly concluding that the outcome must have been what he wanted all along).
Fifth. Partially as a consequence of the previous point, it's probably true that constraints on the president are somewhat less likely to be driven by the sorts of visible things we see in domestic policy. That doesn't mean that those constraints don't exist. It doesn't even mean they won't often show up in visible measures by Congress. It does, perhaps, make it harder to see the constraints in foreign affairs. As a more general caution, I'd point out that fights that reach the level of a Congressional hearing, or a bill on the House or Senate floor, are always a lot easier to see than a cabinet Secretary simply ignoring a presidential request. However, it's the latter that is often the more important constraint on the president. And while all such interactions are difficult for us to see from the outside, secrecy in diplomacy and military operations makes it even harder to guess until decades after the fact whether an American policy proceeds because the president wanted it or over his objections.
I'd close with one of my favorite quotations from Neustadt, but I don't have it well-marked (and I suppose I'm not even 100% certain that I found it there). It's from a cabinet Secretary, who says something to the effect of: the first time a president tells you to do something, it's safe to ignore it. The second time, you need to have an excuse at the ready for why it isn't going to be done. Only when the president tells you a third time do you really have to consider actually doing it. And presidents rarely ask three times.